Re: [PATCH V14] mm/debug: Add tests validating architecture page table helpers

[Christophe Leroy <christophe.leroy@c-s.fr>]

Le 02/03/2020 à 20:40, Qian Cai a écrit :
> On Wed, 2020-02-26 at 10:51 -0500, Qian Cai wrote:
>> On Wed, 2020-02-26 at 15:45 +0100, Christophe Leroy wrote:
>>>
>>> Le 26/02/2020 à 15:09, Qian Cai a écrit :
>>>> On Mon, 2020-02-17 at 08:47 +0530, Anshuman Khandual wrote:
>>>>> This adds tests which will validate architecture page table helpers and
>>>>> other accessors in their compliance with expected generic MM semantics.
>>>>> This will help various architectures in validating changes to existing
>>>>> page table helpers or addition of new ones.
>>>>>
>>>>> This test covers basic page table entry transformations including but not
>>>>> limited to old, young, dirty, clean, write, write protect etc at various
>>>>> level along with populating intermediate entries with next page table page
>>>>> and validating them.
>>>>>
>>>>> Test page table pages are allocated from system memory with required size
>>>>> and alignments. The mapped pfns at page table levels are derived from a
>>>>> real pfn representing a valid kernel text symbol. This test gets called
>>>>> inside kernel_init() right after async_synchronize_full().
>>>>>
>>>>> This test gets built and run when CONFIG_DEBUG_VM_PGTABLE is selected. Any
>>>>> architecture, which is willing to subscribe this test will need to select
>>>>> ARCH_HAS_DEBUG_VM_PGTABLE. For now this is limited to arc, arm64, x86, s390
>>>>> and ppc32 platforms where the test is known to build and run successfully.
>>>>> Going forward, other architectures too can subscribe the test after fixing
>>>>> any build or runtime problems with their page table helpers. Meanwhile for
>>>>> better platform coverage, the test can also be enabled with CONFIG_EXPERT
>>>>> even without ARCH_HAS_DEBUG_VM_PGTABLE.
>>>>>
>>>>> Folks interested in making sure that a given platform's page table helpers
>>>>> conform to expected generic MM semantics should enable the above config
>>>>> which will just trigger this test during boot. Any non conformity here will
>>>>> be reported as an warning which would need to be fixed. This test will help
>>>>> catch any changes to the agreed upon semantics expected from generic MM and
>>>>> enable platforms to accommodate it thereafter.
>>>>
>>>> How useful is this that straightly crash the powerpc?
>>>>
>>>> [   23.263425][    T1] debug_vm_pgtable: debug_vm_pgtable: Validating
>>>> architecture page table helpers
>>>> [   23.263625][    T1] ------------[ cut here ]------------
>>>> [   23.263649][    T1] kernel BUG at arch/powerpc/mm/pgtable.c:274!
>>>
>>> The problem on PPC64 is known and has to be investigated and fixed.
>>
>> It might be interesting to hear what powerpc64 maintainers would say about it
>> and if it is actually worth "fixing" in the arch code, but that BUG_ON() was
>> there since 2009 and had not been exposed until this patch comes alone?
> 
> This patch below makes it works on powerpc64 in order to dodge the BUG_ON()s in
> assert_pte_locked() triggered by pte_clear_tests().
> 
> 
> diff --git a/mm/debug_vm_pgtable.c b/mm/debug_vm_pgtable.c
> index 96dd7d574cef..50b385233971 100644
> --- a/mm/debug_vm_pgtable.c
> +++ b/mm/debug_vm_pgtable.c
> @@ -55,6 +55,8 @@
>   #define RANDOM_ORVALUE	GENMASK(BITS_PER_LONG - 1, S390_MASK_BITS)
>   #define RANDOM_NZVALUE	GENMASK(7, 0)
>   
> +unsigned long vaddr;
> +

Can we avoid global var ?

>   static void __init pte_basic_tests(unsigned long pfn, pgprot_t prot)
>   {
>   	pte_t pte = pfn_pte(pfn, prot);
> @@ -256,7 +258,7 @@ static void __init pte_clear_tests(struct mm_struct *mm,
> pte_t *ptep)
>   
>   	pte = __pte(pte_val(pte) | RANDOM_ORVALUE);
>   	WRITE_ONCE(*ptep, pte);
> -	pte_clear(mm, 0, ptep);
> +	pte_clear(mm, vaddr, ptep);
>   	pte = READ_ONCE(*ptep);
>   	WARN_ON(!pte_none(pte));
>   }
> @@ -310,8 +312,9 @@ void __init debug_vm_pgtable(void)
>   	pgtable_t saved_ptep;
>   	pgprot_t prot;
>   	phys_addr_t paddr;
> -	unsigned long vaddr, pte_aligned, pmd_aligned;

Can we pass local vaddr to pte_clear_tests() instead of making it a 
global var ?

> +	unsigned long pte_aligned, pmd_aligned;
>   	unsigned long pud_aligned, p4d_aligned, pgd_aligned;
> +	spinlock_t *ptl;
>   
>   	pr_info("Validating architecture page table helpers\n");
>   	prot = vm_get_page_prot(VMFLAGS);
> @@ -344,7 +347,7 @@ void __init debug_vm_pgtable(void)
>   	p4dp = p4d_alloc(mm, pgdp, vaddr);
>   	pudp = pud_alloc(mm, p4dp, vaddr);
>   	pmdp = pmd_alloc(mm, pudp, vaddr);
> -	ptep = pte_alloc_map(mm, pmdp, vaddr);
> +	ptep = pte_alloc_map_lock(mm, pmdp, vaddr, &ptl);
>   
>   	/*
>   	 * Save all the page table page addresses as the page table
> @@ -370,7 +373,7 @@ void __init debug_vm_pgtable(void)
>   	p4d_clear_tests(mm, p4dp);
>   	pgd_clear_tests(mm, pgdp);
>   
> -	pte_unmap(ptep);
> +	pte_unmap_unlock(ptep, ptl);
>   
>   	pmd_populate_tests(mm, pmdp, saved_ptep);
>   	pud_populate_tests(mm, pudp, saved_pmdp);
> 

Christophe